PRO-based Stratification Improves Model Prediction for Toxicity and Survival of Head and Neck Cancer Patients

Eric A. Anyimadu; Yaohua Wang; Carla Floricel; Serageldin Kamel; Clifton David Fuller; Xinhua Zhang; G. Elisabeta Marai; Guadalupe M. Canahuate

doi:10.1109/JBHI.2024.3515092

Back

PRO-based Stratification Improves Model Prediction for Toxicity and Survival of Head and Neck Cancer Patients

Journal article

Peer reviewed

PRO-based Stratification Improves Model Prediction for Toxicity and Survival of Head and Neck Cancer Patients

Eric A. Anyimadu, Yaohua Wang, Carla Floricel, Serageldin Kamel, Clifton David Fuller, Xinhua Zhang, G. Elisabeta Marai and Guadalupe M. Canahuate

IEEE journal of biomedical and health informatics, Vol.29(2), pp.807-814

02/2025

DOI: 10.1109/JBHI.2024.3515092

PMCID: PMC11970995

PMID: 40030567

Files and links (1)

url

https://pmc.ncbi.nlm.nih.gov/articles/PMC11970995/View

Open Access

Abstract

Patient-Reported Outcomes (PRO) consist of information provided directly by the patients about their health status including symptom ratings. PROs are commonly used in clinical practice to support clinical decisionmaking and have recently been incorporated into machine learning models to improve risk prediction. In this work, we aim to evaluate whether the inclusion of a patient stratification based on 12-month post-treatment predicted Patient Reported Outcomes improves risk prediction of radiationinduced toxicity and overall survival for head and neck cancer patients. A bidirectional long-short term memory (Bi-LSTM) recurrent neural network was used to model the longitudinal PRO data and to predict symptom ratings 12 months posttreatment. Patients were stratified using hierarchical clustering over the LSTM-predicted data. A logistic regression model was trained to predict Xerostomia at 12 months and a Cox regression model to predict overall survival. Results show that the inclusion of symptom burden clusters derived from the predicted Patient Reported Outcomes improves radiation-induced toxicity and overall survival prediction for head and neck cancer patients.

Bioinformatics

Regression

Toxicology

Biological system modeling

Cancer

Data models

Deep Learning

Long short term memory

Magnetic heads

Neck

Patient Clustering

Patient Reported Outcomes

Predictive models

Survival Analysis

Training

Xerostomia

Details

Title: Subtitle: PRO-based Stratification Improves Model Prediction for Toxicity and Survival of Head and Neck Cancer Patients
Creators: Eric A. Anyimadu - University of Iowa
Yaohua Wang - University of Iowa
Carla Floricel - University of Illinois Urbana-Champaign
Serageldin Kamel - The University of Texas MD Anderson Cancer Center
Clifton David Fuller - The University of Texas MD Anderson Cancer Center
Xinhua Zhang - University of Illinois Urbana-Champaign
G. Elisabeta Marai - University of Illinois Urbana-Champaign
Guadalupe M. Canahuate - University of Iowa
Resource Type: Journal article
Publication Details: IEEE journal of biomedical and health informatics, Vol.29(2), pp.807-814
DOI: 10.1109/JBHI.2024.3515092
PMID: 40030567
PMCID: PMC11970995
NLM abbreviation: IEEE J Biomed Health Inform
ISSN: 2168-2194
eISSN: 2168-2208
Publisher: IEEE
Grant note: NIH: NCI-R01-CA258827
This work was supported by NIH under Award NCI-R01-CA258827.
Language: English
Electronic publication date: 12/09/2024
Date published: 02/2025
Academic Unit: Electrical and Computer Engineering; Internal Medicine
Record Identifier: 9984758287602771

Metrics

11 Record Views

2 Times Cited - Web of Science